Comparative Effects of Insulin-Like Growth Factor II (IGF-II) and IGF-II Mutants Specific for IGF-II/CIM6-P or IGF-I Receptors on in Vitro Hematopoiesis

G.N. Schwartz,a M.K. Warren,b K. Sakano,c J.M. Szabo,a S. W. Kessler,d A. Pashapour,a R.E. Gress,a J.F. Perduee

aTransplantation Therapy Section, National Cancer Institute, Bethesda, Maryland, USA; bOtsuka America Pharmaceutical Company, Rockville, Maryland, USA; cDaiichi Pharmaceutical Company, LTD, Molecular Biology Research Laboratory, Tokyo, Japan; dImmune Cell Biology Program, Naval Medical Research Institute, Bethesda, Maryland, USA; eHolland Laboratories, Molecular Biology Laboratory, American Red Cross, Rockville, Maryland, USA

Key Words. Colony-forming units for granulocytes-macrophages (CFU-GM) • Burst-forming units-erythroid (BFU-E) • Insulin-like growth factor II (IGF-II) • Megakaryocyte • Interleukin 3 • c-kit ligand

Abstract. This report presents the results of stud-ies investigating the effect of insulin‐like growth factor II (IGF‐II) on the proliferation and differen-tiation of CD34+ bone marrow cells in serum‐sub-stituted liquid cultures. Bone marrow cells were enriched for CD34+ cells and then placed in liquid cultures supplemented with either interleukin 3 (IL‐3) or IL‐3 and c‐kit ligand with and without the addition of IGF‐II. When CD34+ cells were incubated with IL‐3, cellularity increased through-out four weeks of culture. Cellularity was twofold greater when cultures also contained IGF‐II. IGF‐II also promoted an increase in cellularity in cultures with IL‐3 and c‐kit ligand. In combina-tion with IL‐3 or IL‐3 and c‐kit ligand, IGF‐II pro-moted an earlier differentiation of granulocytes, as well as an increase in the number of megakaryo-cyte lineage cells. There were approximately two‐fold more colony‐forming units for granulocytes and macrophages (CFU‐GM) and burst‐forming units for erythroid cells (BFU‐E) in cultures con-taining both IL‐3 and IGF‐II than in cultures with IL‐3 alone. These results demonstrate that in cytokine‐supplemented media, physiological con-centrations of IGF‐II augmented both the prolif-eration and differentiation of CD34+ bone marrow cells while maintaining a greater number of pro-genitor cells. To identify the receptors through which IGF‐II enhances in vitro hematopoiesis,

IGF‐II was substituted with one of the mutant forms of IGF‐II that selectively interacts with either IGF‐II/CIM6‐P receptors or with IGF‐I and insulin re-ceptors. The results with the mutant forms of IGF‐II demonstrate that IGF‐II augments in vitro hematopoiesis primarily through its interaction with IGF‐I and possibly insulin receptors, rather than IGF‐ II/CIM6‐P receptors. Stem Cells 1996;14:337-350

Introduction

The mature form of insulin‐like growth factor II (IGF‐II) has a molecular weight of 7500 and is a single‐chain polypeptide of 67 amino acids [1-2]. IGF‐II, alone and in combination with oth-er growth factors, has been shown to be a potent mitogen for a variety of cell lines as well as for cells in primary cultures [1-10]. Results from in vivo and in vitro studies also indicate that IGF‐II is important for stimulating the clonal expansion of embryonic cells and in regulating the growth and differentiation of embryonic and fetal tissues [2, 6, 8-13]. IGF‐II has a high binding affinity for its own receptor, the IGF‐II/cation‐independent man-nose 6‐phosphate (IGF‐II/CIM6‐P) receptor, and also binds to IGF‐I and insulin receptors with moderate to high affinity [1, 2]. The mitogenic effects of IGF‐II on many cell types have been shown to be mediated through the IGF‐I recep-tor [1, 2, 7]. Results from other studies indicate that binding of IGF‐II to IGF‐II/CIM6‐P recep-tors on embryonic or fetal cells may be import-ant in mediating differentiation, in regulating

Correspondence: Dr. Gretchen N. Schwartz, Bldg. 10, Rm. 12N226, Transplantation Therapy Section, Medicine Branch, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892, USA. Received October 16, 1995; provisionally accept-ed December 18, 1995; accepted for publication February 27, 1996. ©AphaMed Press 1066-5099/96/$5.00/0

STEM CELLS 1996;14:337-350

Schwartz, Warren, Sakano et al.

the availability of IGF‐II and in modulating the activation of other proteins that also bind to IGF‐ II/CIM6‐P receptors [1, 2, 10, 13-15]. Others demonstrated that interaction of IGF‐II with IGF‐II/CIM6‐P receptors stimulated the proliferation of some transformed cell lines, as well as mediated motility of human rhabdomyosarcoma cells [11, 16]. The data suggest that differential binding of IGF‐II to the IGF‐I or IGF‐II/CIM6‐P receptors may be important in regulating the proliferation and differentiation of developing tissues. The continuous replacement of mature blood cells and platelets in normal adults is maintained by committed and primitive progenitor cells that are heterogeneous with respect to their self‐ renewal, proliferative and differentiation capacities [17]. In vitro studies demonstrated that physio-logical concentrations of IGF‐II in combination with other growth factors stimulated the prolifer-ation of hematopoietic progenitor cells that form colonies of granulocytes and macrophages (i.e., colony‐forming units for granulocytes and macrophages or CFU‐GM) or erythroid cells (i.e., colony‐forming units for erythroid cells or CFU‐E and burst‐forming units for erythroid cells or BFU‐E) [18-22]. IGF‐II also promoted an increase in the number of CFU‐GM in interleukin 3 (IL‐3) supplemented liquid cultures of peripheral blood cells [23]. In liquid cultures, IL‐3 has been shown to stimulate the proliferation and progressive dif-ferentiation of hematopoietic progenitor cells [24-27]. Liquid cultures of bone marrow cells provide a model to investigate the early effects of IGF‐II on the proliferation and differentiation of hematopoietic progenitor cells and the possible role of the IGF‐I and IGF‐II/CIM6‐P receptors in mediating those responses. The purpose of the studies in this report was to determine the effect of IGF‐II on the clonal expansion, maintenance and differentiation of progenitors for megakaryocytes, granulocytes and macrophages, and erythroid cells. Bone mar-row cells were enriched for hematopoietic pro-genitors by selection of cells expressing the CD34 cell surface antigen. CD34+ cells were placed in liquid cultures supplemented with IL‐3 or IL‐3 and c‐kit ligand, and the effect of the addition or absence of IGF‐II on cellularity, the production of megakaryocyte lineage cells and the number of CFU‐GM and BFU‐E was determined. The results demonstrate that in cytokine‐supple-mented media, IGF‐II enhanced the proliferation and differentiation of granulocyte‐macrophage

and erythroid progenitor cells. To identify the receptors through which IGF‐II enhanced in vitro hematopoiesis, IGF‐II in liquid cultures of CD34+ bone marrow cells was replaced with one of the mutant forms of IGF‐II that binds preferentially to either IGF‐II/CIM6‐P receptors or IGF‐I and insulin receptors [7].

Materials and Methods

Preparation of Cell Suspensions Bone marrow cells were obtained from cadaveric vertebral bodies (LifeNet Transplant Services; Virginia Beach, VA) or from bone mar-row cells aspirated from the iliac crest of nor-mal donors after informed consent under human use protocols approved by the National Cancer Institute and Walter Reed Army Medical Center. Ten to 15 ml of bone marrow aspirate were col-lected in a 20 cc syringe containing preserva-tive‐free heparin (Sigma; St. Louis, MO) as an anticoagulant and then diluted in Dulbecco’s phosphate buffered saline (DPBS) without Ca+2 or Mg+2 (GIBCO; Grand Island, NY) that con-tains the penicillin‐streptomycin (GIBCO). Cells with a density of ≤1.077 g/cm3 were collected and washed three times after separation on a Ficoll‐sodium diatrizoate gradient (Lymphocyte Separation Medium; Organon Teknika Corp.; Durham, NC).

Isolation of CD34+ Bone Marrow Cells A positive immunomagnetic selection pro-cedure was used to select for bone marrow cells that expressed the CD34 antigen [28, 29]. Low density (≤1.077 g/cm3) bone marrow cells were rosetted to magnetic Dynabeads (Dynal Inc.; Great Neck, NY) coated with K6.1, a monoclonal antibody for CD34 and separated by three or four cycles of attraction to a rare earth magnet. CD34+ cells were detached and the beads were magnet-ically separated from the cells. Isolated CD34+

cells were used after overnight incubation in Iscove’s modified Dulbecco’s medium (IMDM), (GIBCO) with 2% fetal bovine serum (FBS). CD34+ cells were also isolated by positive immunomagnetic selection using a high‐gradient magnetic separation column and MiniMacs CD34 progenitor cell isolation kit (Miltenyi Biotec Inc.; Sunnyvale, CA). The purity of the recovered cells was determined by staining with CD34 (anti‐HPCA‐2) phycoerythrin (PE), (Becton Dickinson;

338

Effect of IGF-II on In Vitro Hematopoiesis

San Jose, CA), which binds to a distinct epitope on the CD34 molecule, and flow cytometry analy- sis demonstrated that purity was 95% to 99%. CD34+ cells were suspended in 7.5% dimethyl sulfoxide (DMSO) (Sigma) in 47% FBS and IMDM, cryopreserved in a controlled‐rate freezer and stored in the vapor phase of liquid nitrogen. Cells were quickly thawed in a 37°C water bath, washed twice with excess IMDM, incubated for 1.5 to 2 h at 37°C, 5% CO2 and washed again.

Source of Cytokines and rIGF‐II Recombinant human IL‐3 and c‐kit ligand (R & D Systems; Minneapolis, MN) were resus-pended in DPBS with 0.1% bovine serum albu-min (BSA) and diluted with serum‐substituted medium. Human recombinant IGF‐II, and four mutants of IGF‐II (kindly provided by Daiichi Pharmaceutical, Ltd.; Tokyo, Japan) were pre-pared for use in liquid cultures by placing an aliquot of each in acidified ethanol, drying by speed vac and resuspending to a concentration of 10 μg/ml in 0.1 M acetic acid. Further dilu-tions were prepared in serum‐substituted medium on the day they were added to the liquid cultures. The mutant forms of IGF‐II had been prepared by site‐directed mutagenesis [7]. The two mutants, [Leu43]IGF‐II and [Leu27]IGF‐II were shown to bind with high affinity to the IGF‐II/CIM6‐P receptors and with low affinity to IGF‐I and in-sulin receptors. In contrast, [Arg54,Arg55]IGF‐II and [Thr48,Ser49,Ile50]IGF‐II were shown to bind with high affinity to IGF‐I receptors and with low affinity to IGF‐II/CIM6‐P receptors.

Preparation of Serum‐Substituted Liquid Culture Medium Serum‐substituted liquid culture medium used in these studies was basically as previously described [23]. The basal medium used in these studies was an enriched IMDM supplemented with sodium pyruvate (GIBCO), nonessential amino acids (GIBCO), L‐glutamine (GIBCO), penicil-lin‐streptomycin antibiotic solution (GIBCO) and sodium bicarbonate solution (Quality Bio-logics, Inc.; Gaithersburg, MD) [27]. At the time of culture, IMDM basal medium was further sup-plemented with 20 μg/ml L‐asparagine (Sigma), 1% deionized RIA grade Cohn Fraction V BSA (Sigma) [30], 5 × 10–5 M β‐mercaptoethanol, 300 μg/ml iron‐saturated transferrin, 1% Excyte III (Miles Diagnostic; Kankakee, IL) and 10 μg/ml each of adenosine, cytidine, uridine, guanosine,

2'‐deoxyadenosine, 2'‐deoxycytidine, 2'‐deoxythy-midine and 2'‐deoxyguanosine (all from Sigma). Iron‐saturated transferrin [31] and the riboside mixtures [32] were prepared as 100× solutions, stored at –20°C, and frozen and thawed once. A 1% solution of Excyte III provided 96 to 99 μg/ml cholesterol.

Establishment of Serum‐Substituted Liquid Cultures Liquid cultures of CD34+ bone marrow cells were established in 15 mm Gel‐Well culture chambers (Costar; Cambridge, MA) [23, 27]. In this culture system, close cellular interactions of a small number of cells are maintained and the cells do not adhere to the culture surfaces. A single Gel‐Well was placed in the center of a 35 mm gridded tissue culture dish and 1.5 ml medium was pipetted outside the Gel‐Well. Factors (IL‐3 and IGF‐II) were added to the out-side medium and gently swirled to mix. Either 1 or 2 × 104 CD34+ bone marrow cells were added to the trough inside the Gel‐Well and allowed to settle, then an additional 100 μl of medium were added. In some studies, 6.5 mm nontissue culture‐treated 0.1 μm pore Transwells (Costar) in wells of 24‐well plates were used instead of Gel‐Wells. The outside media and factors were replaced weekly. After one to four weeks of incubation at 37°C, 5% CO2 and 95% room air, the entire contents of individual Gel‐Wells or Transwells were removed, washed and diluted to a final volume of 0.5 ml. Cell concentrations were determined by hemacytometer counts and cell viability was determined by trypan dye exclusion. Slides for cell differential determinations were prepared by placing 1 × 105 cells in 0.1 ml medium with 0.1% BSA into a chamber of a cytocentrifuge (Shandon, Inc.; Pittsburgh, PA) and centrifuging at 800 rpm for 10 min. The slides were air‐dried and then fixed and stained with Hema 3 (Curtin‐Matheson, Inc.; Houston, TX).

Colony Assays for Detection of CFU‐GM and BFU‐E GM‐CFC and BFU‐E determinations were made by subculturing 25 to 100 μl of cells from liquid cultures in 4.0 ml of 1.2% A4M Methocel (Dow Chemical Company; Midland, MI) [30] containing basal IMDM, 1% deionized Fraction V BSA (Sigma) [30], 30% heat inactivated F B S ( H y c l o n e ; L o g a n , U T ) , 5 × 1 0 – 3 M

339

Schwartz, Warren, Sakano et al.

β‐mercaptoethanol (Sigma) and 20 μg/ml L‐asparagine (Sigma). To stimulate colony growth from CFU‐GM, 10% serum‐free PHA‐ LCM (Stem Cell Technologies; Vancouver, Canada) was added to the methylcellulose medium. To stimulate colony growth from BFU‐E, 2 U/ml human recombinant erythro-poietin (R & D Systems) was also added to the colony‐growth medium. One ml of methylcel-lulose medium containing cells was placed in a 35 mm gridded tissue culture dish (Nunc; Naperville, IL) and then maintained in a humid-ified 37°C incubator with 5% CO2 and 95% room air. In one experiment, cells were plated in 0.3% agarose containing 2.5% 10 × 5637 cell‐con-ditioned media [27]. After 14 days, colonies contain-ing ≥50 granulocytes or macrophages or both, or ≥100 erythroid cells were counted as derived from CFU‐GM and BFU‐E, respectively.

Enzyme‐Linked Immunosorbant Assay (ELISA) for Detection of Megakaryocyte Lineage Cells To quantitate production of megakaryo-cytes from CD34+ cells, the cells were placed in liquid culture with serum‐substituted medium containing 2 ng/ml IL‐3 and 20 ng/ml c‐kit lig- and to stimulate proliferation and lineage commitment. After 10 to 11 days of culture, megakaryocyte lineage cells were identified by expression of GP IIb/IIIa (CD41a) and quanti-tated using an enzyme‐linked immunosorbant assay (ELISA) as previously described [33, 34]. The cells were washed three times by centrifu-gation of the 96‐well plates, then all of the cells from each well were transferred to new wells in 96‐well plates that had been pretreated with 1% glutaraldehyde, and centrifuged again. A g lutara ldehyde‐paraformaldehyde mixture (0.004% and 0.1%, respectively) was added for two hours. The wells were rinsed with water, filled with a saline solution containing 0.1% BSA and 0.1% azide, and then the plates were stored at 4°C for up to one week. Quantitation of the megakaryocyte‐specific GP IIb/IIIa and GP Ib antigens was done by ELISA, using the 10E5 and AV6 antibodies, respectively (a generous gift of Dr. Barry Coller, Mt. Sinai, NY). The fixed cells were blocked with 20% goat serum, followed by incubation with primary antibody (10E5 or a negative con-trol antibody). After washing, a biotin‐labeled secondary antibody was added (goat anti‐mouse IgG; Pierce; Rockford, IL) followed by

streptavidin‐conjugated β‐galactosidase (GIBCO). The soluble substrate chlorophenol red‐β‐D‐ga-lactopyranoside (CPRG; Boehringer Mannheim; Indianapolis, IN) was then added. Thirty minutes later, color development was read on a spectro-photometer (Molecular Devices; Menlo Park, CA) at a wavelength of 570 nm. Results are expressed as absorbance (means ± SD of n = 3 or 4 wells). For immunocytochemical stain-ing of positive cells, an insoluble substrate was used following CPRG. The CPRG reac-tion product was washed off, and the substrate bluo‐gal (GIBCO) was added to the ELISA wells. A blue precipitate identified GP IIb/IIIa positive cells [33].

Statistics The paired t‐test was used to test for signif-icant differences in the number of CFU‐GM and BFU‐E in cultures with and without IGF‐II. Paired samples were compared at each time point. The Student’s two‐tailed t‐test was used to test for significant differences within an experiment. Values between different groups were considered to be significantly different for p < 0.05. p val-ues were calculated using Microsoft Excel 4.0 statistical functions.

Results

Effect of IGF‐II on Proliferation and Differentiation of CD34+ Cells in Serum‐ Substituted Liquid Cultures CD34+ bone marrow cells were incubated for one to four weeks in serum‐substituted liquid culture medium. The number of cells decreased in cultures that were not supplemented with IL‐3 (Fig. 1). For example, after two weeks, the num-ber of cells was less than 50% of the number added at day 0. When cultures were supplemented with IL‐3, there was a progressive increase in the number of cells. For example, after two weeks, the cultures contained 3‐ to 11‐fold more cells than added at day 0. IGF‐II alone did not pro-mote an increase in cellularity (data not shown), however, when added to cultures supplemented with IL‐3, 100 and 150 ng/ml of IGF‐II promoted a further twofold increase in cellularity. The number of cells in cultures supplemented with both IL‐3 and c‐kit ligand was greater than in cul-tures with IL‐3 alone (Table 1). Similar to the results with IL‐3, there was a further increase in

340

Effect of IGF-II on In Vitro Hematopoiesis

the number of cells when the cultures were also supplemented with IGF‐II. The stages of granulocytic cells were evalu- ated after the culture of CD34+ cells with IL‐3 alone or with both IL‐3 and IGF‐II. No band or

segmented forms of differentiated granulocytes were detected in the purified preparations of CD34+ cells that were added to the liquid cultures at day 0. After one week, 76% to 88% of the gran-ulocytic cells in cultures with IL‐3 alone and with IL‐3 + IGF‐II were immature (i.e., blast and my-elocyte stages) and 12% to 24% were of the more mature metamyelocyte, band and segmented stages (Fig. 2). After three weeks, the percentage of mature granulocytic cells in cultures supple-mented with both IL‐3 and IGF‐II had already increased to 61%, but was only 47% in cultures with IL‐3 alone. There was also an earlier dif-ferentiation of granulocytic cells when IGF‐II was substituted with [Arg54,Arg55]IGF‐II, a mutant form of IGF‐II with a high affinity to insulin and IGF‐I receptors, but a low affinity for IGF‐II/CIM6‐P receptors.

Effect of IGF‐II on Number of CFU‐GM and BFU‐E in Liquid Cultures of CD34+ Bone Marrow Cells The number of CFU‐GM was determined af-ter the incubation of CD34+ bone marrow cells in liquid cultures with serum‐substituted medium. When CD34+ bone marrow cells were incubat-ed for one and two weeks without growth factors, the number of CFU‐GM was less than 5% of the number added at day 0 (Table 2). There was still a decrease in the total number of CFU‐GM when CD34+ cells were incubated with either IL‐3 alone or with both IL‐3 and c‐kit ligand, however, there were significantly more CFU‐GM than in cultures of cells incubated without the addition of growth factors. In these studies, the decrease in the number of CFU‐GM after one and two weeks was associated with a 1.3‐ to 11‐fold increase in cellularity (Fig. 1).

341

Figure 1. Effect of IL‐3 and IL‐3 plus IGF‐II on the number of cells in serum‐substituted liquid cultures of CD34+ bone marrow cells. Cells (1 × 104) were placed in an enriched serum‐substituted medium containing 2 ng/ml IL‐3 (no insulin or IGF‐II added to the medium), in medium containing IL‐3 plus either 100, or 150 ng/ml IGF‐II, or in medium without either IL‐3 or IGF‐II. Weekly, all of the cells from individual Gel‐ Wells or Transwells were removed, and the number of viable cells per well was determined. Values from cul-tures with no factors are the mean ± SEM from five experiments. Values for cultures with IL‐3 and IL‐3 plus IGF‐II at one and two weeks are the mean ± SEM calculated from nine experiments and at three and four weeks from four and one experiments, respectively. *By paired t‐test, the number of cells was significantly different from the number with IL‐3 alone.

aCD34+ bone marrow cells were incubated for one week, and then all of the cells from each well were collected and the number of viable cells was determined. The values are the mean ± SE values from four different donor marrows. IL‐3 (2 ng/ml), c‐kit ligand (20 ng/ml), and IGF‐II (100 ng/ml) were added to the cultures at day 0.bBy paired t‐test significantly different (p < 0.05) from the number of cells in cultures with IL‐3.cBy paired t‐test significantly different (p < 0.05) from the number of cells in cultures with IL‐3 + c‐kit ligand.

Table 1. Effect of IGF‐II on proliferation of CD34+ bone marrow cells incubated with IL‐3 and c‐kit liganda

Schwartz, Warren, Sakano et al.

The number of CFU‐GM in cultures of CD34+ bone marrow cells incubated with IGF‐II in addition to IL‐3 or a combination of IL‐3 and

c‐kit ligand was also determined. Similar to the response observed with IL‐3 alone, there was a decrease in the total number of CFU‐GM when CD34+ cells were incubated with both IL‐3 and IGF‐II (Fig. 3). Even though there was some heterogeneity in the response of cells from different donors (for example, Figs. 3A and 3B), the number of CFU‐GM in cultures with both IL‐3 and IGF‐II was approximately twofold greater than the number in cultures with IL‐3 alone (Table 3). The number of CFU‐GM in cultures with a combination of IL‐3, c‐kit li-gand and IGF‐II was similar to the number with IL‐3 and c‐kit ligand. A greater number of BFU‐E was also maintained in cultures supple-mented with IGF‐II in addition to IL‐3 (Fig. 4). The decrease in the number of BFU‐E (colonies of ≥100 erythroid cells) in cultures with IGF‐II was associated with an increase in the number of smaller colonies consisting of 20 to 50 erythroid cells that was not observed in cultures with IL‐3 alone. These results demonstrate that the increase in cellularity observed with IGF‐II did not promote an earlier decrease in the number of CFU‐GM or BFU‐E.

Effect of Mutant Forms of IGF‐II on Cellularity and CFU‐GM Numbers in Liquid Cultures of CD34+ Marrow Cells Studies were performed to identify the receptors through which IGF‐II augments in vitro hematopoiesis. In liquid cultures of CD34+

bone marrow cells, one of the mutant forms of IGF‐II that binds with normal or high affinity to either IGF‐IICIM6‐P receptors (i.e.,

342

aCD34+ bone marrow cells were placed in liquid culture with either no factor, 2 ng/ml IL‐3 or IL‐3 and 20 ng/ml c‐kit ligand. One or two weeks later, cells were subcultured in methylcellulose containing PHA‐LCM to stimulate colony growth from CFU‐GMbValues are the mean ± SE calculated from the mean number of colonies from three plates for each experiment.cNumber of different donor marrows.dBy paired t‐test significantly different (p = 0.02) from the number in cultures with IL‐3 alone.eNot determined.

Table 2. Effect of IL‐3 and c‐kit ligand on the number of CFU‐GM in liquid cultures of CD34+ bone marrow cellsa

Figure 2. Effect of IL‐3 and IL‐3 plus IGF‐II on the differentiation of granulocytic cells in serum‐substitut-ed liquid cultures of CD34+ bone marrow cells. Cells were placed in an enriched serum‐substituted medium containing 2 ng/ml IL‐3 (no insulin or IGF‐II added to the medium), in medium containing IL‐3 plus 100 ng/ml IGF‐II or IL‐3 plus 100 ng/ml [Arg54,Arg55]IGF‐II. Weekly, cytospin preparations of cells from individual liquid cultures were prepared and stained. Stages of 100 myeloid cells per culture were determined, and values at one and two weeks were calculated from two liquid cultures and at three weeks from one liquid culture each.

Effect of IGF-II on In Vitro Hematopoiesis

[Leu43]IGF‐II and [Leu27]IGF‐II) or IGF‐I and insulin receptors (i.e., [Arg54,Arg55]IGF‐II and [Thr48,Ser49,Ile50]IGF‐II) was substituted for IGF‐II. Similar to the results observed with

IGF‐II , both [Arg54,Arg55]IGF‐II and [Thr48,S er49,I le50]IGF‐II promoted an increase in cellularity that was greater than with IL‐3 alone (Fig. 5A). In contrast, neither [Leu43]IGF‐II nor [Leu27]IGF‐II promoted an additional increase in cellularity over that observed with IL‐3 alone (Fig. 5B). The num-ber of CFU‐GM in cultures supplemented with a combinat ion of IL‐3 and e ither [Arg54,Arg55]IGF‐II or [Thr48,Ser49,Ile50]IGF‐II was greater than the number in cultures sup-plemented with IL‐3 alone or a combination of IL‐3 and [Leu43]IGF‐II or [Leu27]IGF‐II (Table 4). Similar results were observed for BFU‐E (data not shown). In these studies, neither [Leu27]IGF‐II nor [Leu43]IGF‐II appeared to have an influence on the proliferation and main-tenance of CFU‐GM in cultures of CD34+ bone marrow cells. These results are consistent with the possibility that the increased production or maintenance of CFU‐GM with IGF‐II was medi-ated through the IGF‐I or insulin receptors rather than through the IGF‐II/CIM6‐P receptors.

Effect of IGF‐II and Mutant Forms of IGF‐II on Production of Megakaryocytes Initial studies demonstrated that 6 to 200 ng/ml IGF‐II, [Arg54,Arg55]IGF‐II and [Thr48,Ser49,Ile50]IGF‐II, but not [Leu43]IGF‐II or [Leu27]IGF‐II, enhanced the proliferation of the CMK cell line established from human megakaryoblastic leukemia cells (data not shown). In a preliminary study, when 1 × 104

CD34+ bone marrow cells were incubated for one week with IL‐3 alone there was a 2.2‐fold increase in cellularity. Mature megakaryocytes similar to the large cells with at least four teth-ered nuclei shown in Figure 6 were observed in those cultures. In cultures with both IL‐3 and 100 ng/ml IGF‐II, there were 1.4‐fold more mature megakaryocytes (i.e., 350 with IL‐3 alone versus 500 with IL‐3 + IGF‐II) associated with a similar further increase in total cellularity. In order to determine the effect of IGF‐II on immature and mature megakaryocyte lineage cells produced by CD34+ cells in liquid cultures an ELISA was used to detect platelet glycopro-teins GP IIb/IIIa and GP Ib. CD34+ cells were cultured with IL‐3 and c‐kit ligand with and with-out the addition of IGF‐II. Ten to eleven days later an ELISA was used to detect plate-let glycoproteins GP IIb/IIIa and GP Ib which were expressed as absorbance units. IGF‐II in com-

343

Figure 3. Comparison of the effect of IL‐3 and IL‐3 plus IGF‐II on the number of CFU‐GM in serum‐substituted liquid cultures of CD34+ bone marrow cells. Cells (1 × 104) were placed in an enriched serum‐substituted medium containing 2 ng/ml IL‐3 (no insulin or IGF‐II added to the medium) or in medium containing IL‐3 plus 100 ng/ml IGF‐II. Weekly, cells were subcultured in methylcellulose containing 10% PHA‐LCM for the detection of colonies derived from CFU‐GM. The mean number of colonies from three plates observed after 14 days was used to calculate the total number of CFU‐GM per liquid culture. Values for IL‐3 + IGF‐II are the mean ± SE of the mean values from two Gel‐Well cul-tures or the mean value from one Gel‐Well culture for cells incubated with IL‐3 alone. A and B are the results from two different donor marrows.

Schwartz, Warren, Sakano et al.

bination with IL‐3 and c‐kit ligand promoted an increase in both the total number of cells and the number of megakaryocyte lineage cells that was greater than with IL‐3 and c‐kit ligand alone (Fig. 7). The increase in the number of cells expressing GP IIb/IIIa in response to IGF‐II was also detected by an increase in absorbance values from 0.286 ± 0.019 in cultures without IGF‐II to 1.783 ± 0.136 (mean ± SD for n = 3) in cultures with both IL‐3 and 100 ng/ml of IGF‐II. Absorbance values for expression of GP IIb/IIIa and GP Ib were also increased when either [Arg54,Arg55]IGF‐II or [Thr48,Ser49,Ile50]IGF‐II was substituted for rIGF‐II (Fig. 8). In contrast, neither [Leu27]IGF‐II nor [Leu43]IGF‐II pro-moted an increase in absorbance values that was greater than with IL‐3 and c‐kit ligand alone. These results are consistent with the possibility that the increased production of megakaryocyte lineage cells with IGF‐II was mediated through the IGF‐I or insulin receptors rather than through the IGF‐II/CIM6‐P receptors.

Discussion

The studies in this report investigated the effect of IGF‐II on cellularity and the number of committed myeloid and erythroid progenitor cells in serum‐substituted liquid cultures of CD34+

bone marrow cells. In these studies, IL‐3 alone (i.e., no IGF or insulin added to the medium) or IL‐3 plus c‐kit ligand promoted the proliferation of CD34+ bone marrow cells. When the cultures were also supplemented with IGF‐II there was a further twofold increase in the total number of cells. IGF‐II also promoted an increase in the

344

aCD34+ bone marrow cells were placed in liquid culture. One or two weeks later, cells were subcultured in methylcel-lulose containing PHA‐LCM to stimulate colony growth from CFU‐GM. At day 0, the cultures were supplemented with either 2 ng/ml IL‐3, IL‐3 and 100 ng/ml IGF‐II, IL‐3 and 20 ng/ml c‐kit ligand, or a combination of IL‐3, c‐kit ligand, and IGF‐II.bValues are the mean ± SE calculated from the mean number of colonies from three plates for each experiment.cNumber of different donor marrows.dBy paired t‐test significantly different (p < 0.02) from the number in cultures with IL‐3 alone.eNot determined.

Table 3. Effect of IGF‐II on the number of CFU‐GM in liquid cultures of CD34+ bone marrow cellsa

Figure 4. Comparison of the effect of IL‐3 and IL‐3 plus IGF‐II on the number of BFU‐E in serum‐sub-stituted liquid cultures of CD34+ bone marrow cells. Cells (1 × 104) were placed in an enriched serum‐sub-stituted medium containing 2 ng/ml IL‐3 (no insulin or IGF‐II added to the medium) or in medium contain-ing IL‐3 plus 100 ng/ml IGF‐II. One and two weeks later, cells were subcultured in methylcellulose con-taining 10% PHA‐LCM and 2 U/ml erythropoietin for the detection of colonies derived from BFU‐E (≥100 erythroid cells) and small erythroid colonies (20 to 50 cells). The mean number of colonies from three plates observed after 14 days was used to calculate the total number of colonies per liquid culture. Solid lines are values from cultures supplemented with IL‐3 alone and dashed lines are values from cultures supplemented with IL‐3 and IGF‐II. Values for IL‐3 + IGF‐II are the mean ± SE of the mean values from two Gel‐Well cul-tures or the mean value from one Gel‐Well culture for IL‐3 alone. The data for this figure are from the same experiment as shown in Figure 3B.

Effect of IGF-II on In Vitro Hematopoiesis

number of megakaryocyte lineage cells and an increase in the proportion of mature differentiated granulocytes. The increase in cellularity with the addition of IL‐3 alone was associated with a decrease in the number of myeloid and erythroid colony‐forming cells. The additional increase in cellularity observed with IL‐3 and IGF‐II was not associated with a further reduction in the number of colony‐forming cells. These results suggest that IGF‐II supported the maintenance of a greater number of committed progenitor cells. To identify the receptors through which IGF‐II augmented in vitro hematopoiesis, one of the mutant forms of IGF‐II that has a high affin-ity for either IGF‐II/CIM6‐P receptors or IGF‐I and insulin receptors and a reduced affinity for the heterologous receptor was substituted for IGF‐II [7]. The results indicate that the increase in cellularity and differentiation observed in cul-tures with IGF‐II was mediated primarily through the IGF‐I or insulin receptors rather than through the IGF‐II/CIM6‐P receptors. IGF‐II has been shown to be important in promoting the proliferation and differentiation of fetal and embryonic cells including myoblasts, metanephros and transplanted embryos [2, 6, 8-10, 12-14, 35]. Others demonstrated that 0.2 to 1000 ng/ml IGF‐II increased the number of erythroid and granulocyte‐macrophage colonies formed in serum‐substituted colony growth medium con-taining saturating concentrations of erythropoi-etin or other growth factors [18-22]. Colony size was also increased in cultures supplemented with IGF‐II [21-22]. A previous report demonstrated that IGF‐II increased IL‐3‐induced expansion of CFU‐GM in serum‐substituted liquid cultures of peripheral blood cells [23]. The present studies were expanded to investigate the effect of IGF‐II on the proliferation and differentiation of granu-locyte‐macrophage, erythroid and megakaryocyte progenitors from CD34+ bone marrow cells. In these studies, IGF‐II alone (i.e., no other growth factors were added to the medium) had no effect on the number of cells in liquid cultures of CD34+ bone marrow cells. In contrast, IGF‐II amplified the stimulatory effects of IL‐3 and c‐kit ligand on cellularity and differentiation. With IL‐3 alone, cellularity increased throughout four weeks of culture and there was also a progres-sive development of mature band and segmented stages of granulocytes from the more immature blast and myelocyte stages. When the cultures were also supplemented with IGF‐II, there was

345

Figure 5. Effect of mutant forms of IGF‐II on the number of cells in serum‐substituted liquid cultures of CD34+ bone marrow cells. Cells (1 × 104) were placed in an enriched serum‐substituted medium con-taining 2 ng/ml IL‐3 (no insulin or IGF‐II added to the medium), in medium containing IL‐3 plus 100 or 150 ng/ml of one of the mutant forms of IGF‐II. For individual experiments the same concentration was used for all four mutant forms of IGF‐II. Weekly, all of the cells from Gel‐Wells or Transwells were removed, and the number of viable cells per well was determined. The number of cells recovered from cul-tures supplemented with mutant forms of IGF‐II was compared to the number in cultures with IL‐3 alone. IGF‐II was substituted with: A) mutant forms of IGF‐II with a normal to high affinity for the IGF‐I and insulin receptors and a reduced affinity for IGF‐ II/CIM6‐P receptors or B) mutant forms of IGF‐II with normal affinity for IGF‐II/CIM6‐P receptors and a reduced affinity for IGF‐I receptors. Values are the mean ± SE calculated from seven experiments for one and two weeks and from three experiments at three weeks. *By paired t‐test, the number of cells was significantly different (p ≤ 0.04) from the number with IL‐3 alone.

Schwartz, Warren, Sakano et al.

a further twofold increase in total cellularity, and a greater proportion of granulocytic cells were of the mature band and segmented stages. Megakaryocyte lineage cells were detectable when CD34+ bone marrow cells were stimulated with IL‐3 alone or IL‐3 plus c‐kit ligand, however, there was an increased number of megakaryocytes in cultures also supplemented with IGF‐II. Results of studies by Debili et al. [36] indicated that megakaryocyte lineage cells express the platelet glycoprotein GP IIb/IIIa during early stages

of differentiation and that GP Ib is expressed at more mature stages of megakaryocyte de-velopment. In the present studies, there was an increase in the expression of both GP IIb/IIIa and GP Ib. These results suggest that, in ad-dition to stimulating an increase in the pro-duction of megakaryocytes, IGF‐II also had stimulatory effects on megakaryocyte mat-uration. There were also more CFU‐GM and BFU‐E in cultures with both IGF‐II and IL‐3 than in cultures with IL‐3 alone. The results of the present studies demonstrate that in cytokine‐supplemented media, IGF‐II augmented both the proliferation and differentiation of CD34+ bone marrow cells while maintaining a greater number of committed progenitor cells. IGF‐II binds with moderate to high affini-ty to IGF‐II/CIM6‐P, IGF‐I and insulin receptors [1, 2]. The mitogenic effects of IGF‐II on many types of cells have been shown to be mediated through the IGF‐I receptor [1, 2, 7]. For exam-ple, studies by Merchav et al. [21, 22] demon-strated that the stimulatory effects of IGF‐II on myeloid and erythroid colony formation were prevented when IGF‐I receptors were blocked with the anti‐IGF‐I receptor antibody, αIR‐3. The results of studies with K562 erythro-leukemia cells, however, indicate that the binding of IGF‐II to IGF‐II/CIM6‐P receptors may also stimulate cell proliferation [11]. In the present studies, the effect of the mutant

346

aCD34+ bone marrow cells were placed in liquid culture. One or two weeks later, cells were subcultured in methylcel-lulose containing PHA‐LCM to stimulate colony growth from CFU‐GM. At day 0, the cultures were supplemented with 2 ng/ml IL‐3 and 100 or 150 ng/ml of one of the mutant forms of IGF‐II.bValues are the mean ± SE calculated from the mean number of colonies from three plates for each experiment.cNumber of different donor marrows.dBy paired t‐test significantly different (p < 0.05) from the number in cultures with IL‐3 alone.

Table 4. Effect of mutant forms of IGF‐II on the number of CFU‐GM in liquid cultures of CD34+ marrow cellsa

Figure 6. Photomicrograph of mature megakaryo-cyte lineage cells similar to those detected in liquid cultures of CD34+ bone marrow cells incubated with IL‐3 and both IL‐3 and IGF‐II. Figure is from 100× magnification.

Effect of IGF-II on In Vitro Hematopoiesis

forms of IGF‐II on the proliferation of CD34+

bone marrow cells was investigated. Analogs of IGF‐II prepared by site‐directed mutagenesis were designed to selectively interact with either IGF‐I or IGF‐II/CIM6‐P receptors and have a greatly reduced affinity for the other receptor type [7]. For example, [Leu27]IGF‐II and [Leu43]IGF‐II have a normal affinity for the IGF‐II/CIM6‐P receptor, but an 80‐ to 220‐fold reduced affinity for the IGF‐I and insulin recep-tors. In contrast, [Thr48,Ser49,Ile50]IGF‐II and [Arg54,Arg55]IGF‐II have normal to high affinities for the IGF‐I and insulin receptors, but low or no affinity for the IGF‐II/CIM6‐P receptors. Both [Arg54,Arg55]IGF‐II and [Thr48,Ser49,Ile50]IGF‐II promoted an increase in cellularity that was similar to that of recombi-nant IGF‐II. In contrast, neither [Leu27]IGF‐II nor [Leu43]IGF‐II promoted an increase in cel-lularity that was greater than with IL‐3 alone. These results along with those of Merchav et al. [21, 22] are consistent with the possibility that the stimulatory effects of IGF‐II on the pro-liferation of CD34+ bone marrow cells were mediated through the IGF‐I or insulin receptors rather than through the IGF‐II/CIM6‐P. Result by Rogers et al. [10] and Rosenthal et al. [13] demonstrated that in vitro differen-tiation of developing renal anlage and smooth muscle, respectively, was mediated through IGF‐II/CIM6‐P receptors [10, 13]. In the present studies, IGF‐II promoted an increase in the pro-duction of megakaryocytes from CD34+ bone marrow cells that was detected by an increase

in expression of the platelet glycoproteins GP IIb/IIIa and GP Ib. A similar increase in platelet

347

Figure 7. Immunocytochemical staining for the detection of GP IIb/IIIa positive cells. The photomicrographs show that when compared to IL‐3 plus c‐kit ligand alone (A), 100 ng/ml IGF‐II promoted an increase in cellularity and in the number of cells positive for GP IIb/IIIa (B) that corresponded to an increase in absorbance.

Figure 8. Effect of mutant forms of IGF‐II on the production of megakaryocytes from CD34+ bone mar-row cells. Cells (4 × 103) were placed in wells of 96‐well plates containing an enriched serum‐substituted medium containing 2 ng/ml IL‐3 plus 20 ng/ml c‐kit ligand (no insulin or IGF‐II added to the medi-um) or in medium containing IL‐3 plus 100 ng/ml of [Arg54,Arg55]IGF‐II (A), [Thr48,Ser49,Ile50]IGF‐II (B), [Leu27]IGF‐II (C), or [Leu43]IGF‐II (D). Eleven days later the cells were fixed and stained for platelet glycoproteins GP IIb/IIIa and GP Ib. The expression of GP IIb/IIIa and GP Ib was measured by ELISA and expressed as absorbance units. The values are the mean ± SD from three wells.

Schwartz, Warren, Sakano et al.

glycoprotein expression was observed when IGF‐II was substituted with either [Arg54,Arg55]IGF‐II or [Thr48,Ser49,Ile50]IGF‐II, but not with either [Leu27]IGF‐II or [Leu43]IGF‐II. Results in the present studies suggest that in addition to pro-moting an increase in megakaryocyte production and maturation, [Arg54,Arg55]IGF‐II also increased the differentiation rate of granulocytes. These results suggest that stimulatory effects of IGF‐II on the differentiation of granulocyte and megakaryocyte lineage cells was not mediated through IGF‐II/CIM6‐P receptors, but by binding of IGF‐II to IGF‐I or insulin receptors. The results in the present studies demon-strate that physiological concentrations of IGF‐II augmented cytokine‐induced proliferation and differentiation of CD34+ bone marrow cells. These results and those by others [18-23] sug-gest that in addition to its role in fetal develop-ment, IGF‐II may have a role in the regulation of adult hematopoiesis. It is not clear whether the stimulatory effects of IGF‐II on in vitro hematopoiesis are in response to direct effects on CD34+ progenitor cells or to effects of IGF‐II on other cells that are produced in response to IL‐3 and c‐kit ligand. In other studies, the cell sus-pensions were not necessarily depleted of acces-sory cells such as monocytes and lymphocytes or of relatively immature committed cells that may have responded to IGF‐II. In the present studies, bone marrow cells with a relatively high purity of CD34+ cells (≥95%) were used, how-ever, studies of the effects of IGF‐II on cell pro-liferation were not performed until at least one week of culture. During that time, IL‐3 alone pro-moted cell proliferation and differentiation. IGF‐II may have enhanced further differentiation or acted as a survival factor for cells produced in response to other stimulatory factors. Single cell experi-ments and studies at earlier times will be import-ant in delineating the direct and indirect effects of IGF‐II on CD34+ cells. Multiple pathways appear to be important in regulating the biological func-tions and gene expression of IGF‐II. Some effects of IGF‐II are related to direct interactions of IGF‐II with IGF‐I or IGF‐II/CIM6‐P receptors [1, 2, 7, 10, 13, 14, 35] or regulated by its interac-tion with IGF‐binding proteins [2]. IGF‐II can also regulate the gene expression of IGF‐II [12, 35], the expression of IGF‐I receptors [12], the cell surface expression of IGF‐II/CIM6‐P re-ceptors [37], and act as a second signal for pro-liferation [38]. The importance of these other

aspects of IGF‐II function and regulation on hema-topoiesis are not known.

Acknowledgements

We are grateful to Mr. Alfred Black and Ms. Mary Cutting for their skillful technical as-sistance. This research was supported in part by Naval Medical Research and Development Command work unit number 63105A DH29 AD 010. Portions of this work were done while G.N. Schwartz, Ph.D., held a Senior National Research Council Research Associateship at Walter Reed Army Institute of Research. Disclaimer: The opinions or assertions contained herein are the private views of the authors and are not to be construed as official or reflecting the views of the Departments of the Army or Navy or the Department of Defense.

References

348

1 Humbel RE. Review: insulin-like growth factors I and II. Eur J Biochem 1990;190:445-462.

2 Nielsen FC. The molecular and cellular biology of insulin-like growth factor II. Prog in Growth Factor Res 1992;4:257-290.

3 Hampton B, Burgess WH, Marshak DR et al. Purification and characterization of an insulin- like growth factor II variant from human plasma. J Biol Chem 1989;264:19155-19160.

4 Gowan LK, Hampton B, Hill DJ et al. Purification and characterization of a unique high molecu-lar weight form of insulin-like growth factor II. Endocrinol 1987;121:449-458.

5 Oksenberg D, Dieckmann BS, Greenberg PL. Functional interactions between colony-stimu-lating factors and the insulin family hormones for human myeloid leukemic cells. Cancer Res 1990;50:6471-6477.

6 Werther GA, Haynes K, Johnson GR. Insulin-like growth factors promote DNA synthesis and support cell viability in fetal hemopoietic tissue by paracrine mechanisms. Growth Factors 1990;3:171-179.

7 Sakano K, Enjoh T, Numata F et al. The design, expression, and characterization of human insulin- like growth factor II (IGF-II) mutants specific for either the IGF-II/cation-independent mannose 6-phosphate receptor or IGF-I receptor. J Biol Chem 1991;266:20626-20635.

Effect of IGF-II on In Vitro Hematopoiesis349

20 Sander M, Sorba S, Dainiak N. Insulin-like growth factors stimulate erythropoiesis in serum- substituted umbilical cord blood cultures. Exp Hematol 1993;21:25-30.

21 Merchav S, Silvian-Drachsler 1, Tatarsky I et al. Comparative studies of erythroid-potentiating effects of biosynthetic human insulin-like growth factors I and 11. J Clin Endocrinol Metabol 1992;74:447-452.

22 Merchav S, Mats M, Skottner A. Comparative studies of the granulopoietic enhancing effects of biosynthetic human insulin-like growth fac-tors I and II. J Cell Physiol 1993;157:178-183.

23 Schwartz GN, Hudgins WR, Perdue JF. Glycosylated insulin-like growth factor II pro-moted expansion of granulocyte-macrophage colony-forming cells in serum-deprived liquid cul-tures of human peripheral blood cells. Exp Hematol 1993;21:1447-1454.

24 Saeland S, Caux C, Favre C et al. Effects of recombi-nant human interleukin-3 on CD34-enriched nor-mal hematopoietic progenitors and on myeloblastic leukemia cells. Blood 1988;72:1580-1588.

25 Kobayashi M, Van Leeuwen BH, Elsbury S et al. lnterleukin-3 is significantly more effective than other colony-stimulating factors in long-term main-tenance of human bone marrow-derived colony- forming cells in vitro. Blood 1989;73:1836-1841.

26 Saeland S, Duvert V, Caux C et al. Distribution of surface-membrane molecules on bone marrow and cord blood CD34+ hematopoietic cells. Exp Hematol 1992;20:24-33.

27 Schwartz GN. Use of Gel-Well culture chambers as a liquid culture system to measure responses of hematopoietic colony-forming cells to growth factors. Int J Cell Cloning 1989;7:360-372.

28 Folks TM, Kessler SW, Orenstein JM et al. In-fection and replication of HIV-1 in purified progenitor cells of normal human bone marrow. Science 1988;242:919-922.

29 Schwartz GN, Kessler SW, Szabo JM et al. Negative regulators may mediate some of the inhibitory effects of HIV-1 infected stromal cell layers on erythropoiesis and myelopoiesis in hu-man bone marrow long term cultures (LTC). J Leuk Biol 1995;57:948-955.

30 Worton RG, McCulloch EA, Till JE. Physi-cal separation of hemopoietic stem cells from cells forming colonies in culture. J Cell Physiol 1969;74:171-182.

31 Iscove NN. Culture of lymphocytes and hematopoietic cells in serum-free medium. In: Barnes DW, Sibasku DA, Sato GH, eds. Methods for Serum-Free Culture of Neuronal and Lymphoid Cells. New York, NY: Liss, 1984:169-185.

8 Liu L, Greenberg S, Russell SM et al. Effects of insulin-like growth factors I and II on growth and differentiation of transplanted rat embryos and fetal tissues. Endocrinol 1989;124:3077-3082.

9 Vetter U, Zapf J, Heit W et al. Human fetal and adult chondrocytes: effect of insulin-like growth factors I and II, insulin and growth hormone on clonal growth. J Clin Invest 1986;77:1903-1908.

10 Rogers SA, Ryan G, Hammerman MR. Insulin- like growth factors I and II are produced in the metanephros and are required for growth and de-velopment in vitro. J Cell Biol 1991;113:1447-1453.

11 Tally M, Li CH, Hall K. IGF-2 stimulated growth mediated by the somatomedin type 2 receptor. Biochem Biophys Res Commun 1987;148:811-816.

12 DeChiara TM, Efstratiadis A, Robertson EJ. A growth-deficiency phenotype in heterozygous mice carrying an insulin-like growth factor II gene disrupted by targeting. Nature 1990;345:78-80.

13 Rosenthal SM, Hsiao D, Silverman LA. An insulin-like growth factor-II (IGF-II) analog with highly selective affinity for IGF-Il receptors stimulates differentiation, but not IGF-I receptor down-regulation in muscle cells. Endocrinol 1994;135:38-44.

14 Florini JR, Ewton DZ, Falen SL et al. Biphasic concentration dependency of the stimulation of myoblast differentiation by somatomedins. Am J Physiol 1986;250:c771-c778.

15 Murayama Y, Okamoto T, Ogata E et a l . Distinctive regulation of the functional linkage between the human cation-independent mannose 6-phosphate receptor and GTP-binding proteins by insulin-like growth factor II and mannose 6 -phosphate. J Biol Chem 1990;265:17456-17462.

16 Minniti CP, Kohn EC, Grubb JH et al. The insulin-like growth factor 11 (IGF-II)/mannose 6-phosphate receptor mediates IGF-II-induced motility in human rhabdomyosarcoma cells. J Biol Chem 1992;267:9000-9004.

17 Sutherland HJ, Eaves AC, Eaves CJ. Quantitative assays for human hemopoietic progenitor cells. In: Gee AP, ed. Bone Marrow Processing and Purging. A Practical Guide. Boca Raton, FL: CRC Press, 1991:155-171.

18 Dainiak N, Kreczko S. Interactions of insulin, insulin-like growth factor 11, and platelet-derived growth factor in erythropoietic culture. J Clin Invest 1985;76:1237-1242.

19 Dainiak N, Sanders M, Sorba S. Induction of cir-culating neonatal stem cell populations. Blood Cells 1991;17:339-343.

Schwartz, Warren, Sakano et al. 350

growth factor-II gene expression in differentiating myoblasts in vitro. Endocrinol 1994;135:53-62.

36 Debili N, lssaad C, Masse J-M et al. Expres-sion of CD34 and platelet glycoproteins during human megakaryocytic differentiation. Blood 1992;80:3022-3035.

37 Damke H, von Figura K, Braulke T. Simultaneous redistribution of mannose 6-phosphate and trans-ferrin receptors by insulin-like growth factors and phorbol ester. Biochem J 1992;281:225-229.

38 Christofori G, Naik P, Hanahan D. A second signal supplied by insulin-like growth factor 11 in oncogene-induced tumorigenesis. Nature 1994;369:414-418.

32 Migliaccio G, Migliaccio AR. Cloning of human erythroid progenitors (BFU-E) in the absence of fetal bovine serum. Br J Haematol 1987;67:129-133.

33 Warren MK, Guertin M, Rudzinki I et al. A new culture and quantitation system for megakaryocyte growth using cord blood CD34+ cells and the GP Ilb/Illa marker. Exp Hematol 1993;21:1473-1479.

34 Warren MK, Rose WL, Beall LD et al. CD34+ cell expansion and expression of lineage markers during liquid culture of human progenitor cells. STEM CELLS 1995;13:167-174.

35 Magri KA, Benedict MR, Ewton DZ et al. Negative feedback regulation of insulin-like